Multifunctional S-nitroso-N-acetylpenicillamine Incorporated Medical Grade Polymer with Selenium Interface for Biomedical Applications

Abstract

Modern crisis in implantable or indwelling blood-contacting medical devices are mainly due to the dual problems of infection and thrombogenicity. There is a paucity of biomaterials that can address both the problems simultaneously through a singular platform. Taking cues from the body’s own defense mechanism against infection and blood clotting (thrombosis) via the endogenous gasotransmitter nitric oxide (NO), both of these issues are addressed through the development of a layered S-nitroso-N-acetylpenicillamine (SNAP) doped polymer with a blended selenium (Se) - polymer interface. The unique capability of the SNAP-Se-1 polymer composites to explicitly release NO from the SNAP reservoir as well as generate NO via the incorporated Se is reported for the first time. The NO release from the SNAP doped polymer increased substantially in the presence of the Se interface. The Se interface is also able to generate NO in the presence of S-nitrosoglutathione (GSNO) and glutathione (GSH), demonstrating the capability of generating NO from endogenous S-nitrosothiols (RSNO). Scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS) traces distribution of elemental Se nanoparticles on the interface and the surface properties were evaluated by surface wettability and roughness. The SNAP-Se-1 efficiently inhibits the growth of bacteria and reduces platelet adhesion while showing minimal cytotoxicity, thus potentially eliminating the risks of systemic antibiotic and blood coagulation therapy. The SNAP-Se-1 exhibits antibacterial activity of ~2.39 and ~2.25 log reductions in the growth of clinically challenging adhered Gram-positive Staphylococcus aureus and Gram-negative Escherichia coli. SNAP-Se-1 also significantly reduces platelet adhesion by 85.5% compared to corresponding controls. WST-8 based cell viability test performed on NIH 3T3 mouse fibroblast cells provide supporting evidence for the potential biocompatibility of the material in vitro. These results highlight the prospective utility of SNAP-Se-1 as a blood-contacting infection-resistant biomaterial in vitro which can be further tuned by application specificity.

Graphical Abstract

1. Introduction

Indwelling or implantable medical devices such as stents, orthopedic implants, catheters, heart valves, and vascular grafts have revolutionized the way patients are treated for various diseases. However, increasing incidences of medical device failure have posed a serious problem, leading to increased rates of patient mortality and morbidity. Despite advancements in the materials and design of implantable medical devices, two persistent problems that contribute to device failure and limit successful therapeutics are bacterial infection and thrombus formation.^1-2

Medical device-associated infections account for 47.4% of hospital acquired infections (HAIs) in critically ill patients.³ At least 700,000 patients are affected annually by HAIs, leading to a staggering financial burden of approximately $45 billion in US alone.^4-5 The rapid emergence of antibiotic resistance has aggravated nosocomial infections and is one of the major driving forces behind reducing efficacy of antibiotic therapy in treating such infections.⁶ Attachment of planktonic bacteria on biomaterial surface marks the onset of bacterial colonization. Pathogenic bacteria physically aggregate on the surface and eventually forms a clustered ecosystem surrounded by a polymer matrix called biofilms.⁷ Biofilms play a pivotal role in the infection of implantable medical devices by drastically reducing the susceptibility of antimicrobial agents through intrinsic or acquired (e.g. plasmid exchange etc.) mechanisms.⁸ Most clinically relevant bacteria such as Staphylococcus aureus (S. aureus), Escherichia coli (E. coli), Pseudomonas aeruginosa (P. aeruginosa), Enterococcus faecalis, and Staphylococcus epidermidis (S. epidermis) are associated with biofilm formation on medical devices. Diffusion of antimicrobial agents through the extracellular polysaccharides of a biofilm is significantly reduced by limiting their rate of transport.⁸ Due to increased tolerance to antibiotics and resistance towards phagocytosis and other components of the body's defense system, bacterial biofilms can result in chronic infections.^9-10

For blood-contacting devices, exposure to blood can activate complex, intermingled processes leading to protein adsorption and platelet adhesion, ultimately resulting in device occlusion and clot formation.^11-12 Medical device-induced thrombosis severely limits the success of some of the most widely utilized life-saving cardiovascular and renal treatments. In fact, device-induced clotting has been reported to occur in 22% of extracorporeal life support treatments.¹³ Consequently, blood-contacting devices require the use of systemic anti-platelet or anti-coagulant therapy; however, the administration of anti-platelet drugs, such as clopidogrel, or anticoagulants, such as unfractionated heparin, can often lead to increased or uncontrolled bleeding and heparin-induced thrombocytopenia.^14-15 Moreover, thrombus formation arising from interaction between blood-contacting device interfaces and platelets may increase the likelihood of biofilm formation, as the adhered plasma proteins aid in bacterial adhesion.¹⁶ Platelets adhering to surfaces can play a role in pathogenesis of bacteria like S. aureus in causing bloodstream infections.¹⁷

Despite extensive research, an ideal non-thrombogenic and infection-resistant biomaterial remains an open challenge in the field of medical devices. The biosynthesis of NO within endothelial cells, macrophages, and endogenous RSNOs in blood play an important role in inhibiting platelet adhesion and aggregation to the blood vessels and providing innate immunity against bacterial infections.^18-19 The short half-life (on the order of seconds), rapid diffusion and reactivity of NO can help reduce a broad range of microbial burden.²⁰ Exogenous delivery of NO via NO donor molecules have been previously shown to be effective against HAI-relevant strains of bacteria such as S. aureus, E. coli, S. epidermis, and P. aeruginosa, making it an attractive candidate for improving antimicrobial activity.²¹ To improve antibacterial and antiplatelet activity of biomaterials, both exogenous and endogenous NO donors such as RSNOs, N-diazeniumdiolates, organic nitrates and nitrites, and metal nitrosyl complexes have been utilized to provide localized NO release to surrounding tissue.^21-22 RSNOs, among others, have emerged as exceptionally beneficial due to their steady release and innate biocompatibility.²³

However, exogenous mechanisms of NO delivery in blood contacting devices have been limited to either NO-release (NOrel) through donor molecules or NO-generation (NOgen) catalytically generating NO at the device interface via interaction with endogenous RSNOs in the blood circulation.^24-25 NOrel materials rely on a finite reservoir of donor molecules which release NO at favorable physiological conditions. In contrast, NOgen materials catalytically generate NO at local material interfaces, often involving natural enzymes or catalysts such as metal ions (Cu²⁺, Fe²⁺, Co²⁺, Ni²⁺, Zn²⁺).²⁶ Either approach by itself has some shortcomings. NOrel strategy which relies on a finite source of NO suffer from donor exhaustion over time, which in turn impairs longevity of NOrel materials. While NOgen polymers involving catalytic NO generation via circulating RSNO has the potential to solve those issues, the initial NO flux of NOgen materials may not be adequate in preventing early onslaught of infection or platelet activation. In a 4 h experiment on rabbit thrombogenicity model by Major et al., although extracorporeal loops coated with 10 wt% Cu combined with a saline infusion improved platelet counts, loops coated with 10 wt% Cu with a systemic SNAP infusion best preserved platelet counts.²⁷ However, systemic SNAP infusion can cause hypotension²⁸, hyperglycemia²⁹ and/or decrease in cell viability.³⁰ The current deficits can be solved through a robust combination which integrates properties of NOrel and NOgen in a single polymer composite, simultaneously addressing the issues of infection and thrombosis, as well as preventing suppressive antibiotic or antiplatelet therapies.

In the physiological production of NO, Se-based enzymes or organoselenium species facilitate endogenous generation of NO by catalytically decomposing RSNO.^31-33 Selenium, an essential trace element, plays a crucial function as the active site of numerous Se-dependent enzymes such as peroxidase, iodothyronine deiodinase, glutathione peroxidase (GPx).³⁴ It has been established that diselenides catalyze the decomposition of RSNOs in the presence of thiols to produce NO, and GPx alone can catalyze the same without the need for thiols or hydrogen peroxide (H₂O₂).³⁵ Se nanoparticles have also been noted to have innate antimicrobial and platelet-modulating capabilities when doped with other polymers or inorganic compounds.^36-37

In this study, the unique combination of NOrel and NOgen chemistries within a medical grade polymer has been assessed. The combination can elevate the initial NO release from SNAP reservoir as well as sustain long-term release from endogenous RSNOs which can be instrumental in averting challenges of infection and thrombosis associated with device implants. After confirming the distribution and surface morphology via SEM-EDS and contact angle studies, the antibacterial efficacy, anti-platelet adhesion, and cytotoxic properties of polymer composites were studied in vitro.

2. Materials and methods

2.1. Materials

N-Acetyl-D-penicillamine (NAP), tetrahydrofuran (THF), sodium nitrite (NaNO₂), ethylenediaminetetraacetic acid (EDTA), hydrochloric acid, and selenium powder (−100 mesh) were purchased from Sigma Aldrich (St. Louis, MO 63103). CarboSil-2080A™ (CarboSil) was purchased from DSM (Berkeley, CA). Phosphate-buffered saline (PBS), pH 7.4, used for in vitro experiments contained 138 mM NaCl, 2.7 mM KCl, and 10 mM sodium phosphate. Dulbecco’s modified Eagle’s medium (DMEM) and trypsin-EDTA were purchased from Corning (Manassas, VA 20109). The Cell Counting Kit-8 (CCK-8) was purchased from Sigma Aldrich (St. Louis, MO 63103). Penicillin-Streptomycin (Pen-Strep) and fetal bovine serum (FBS) was obtained from Gibco-Life Technologies (Grand Island, NY 14072). The bacterial strains S. aureus (ATCC 6538) and E. coli (ATCC 11303) were purchased from American Type Culture Collection (ATCC). LB broth was obtained from Fisher Bioreagents (Fair Lawn, NJ). LB Agar was purchased from Difco Laboratories Inc (Detroit, MI). Reagents used for NO generation. The lactic dehydrogenase (LDH) kit was purchased from Roche Life Sciences (Indianapolis, IN). Glutathione reduced was purchased from Goldbio (St. Louis, MO).

2.2. Synthesis of SNAP

SNAP was synthesized from NAP by modification of a previously established protocol.³⁸ Equimolar ratio of NaNO₂ and NAP were added to a mixture of de-ionized water and methanol containing 1 M HCl and 1 M H₂SO₄. The mixture was stirred in a reaction vessel in the absence of light (to avoid activation of NO release with light as a stimulant) for 15 minutes and then cooled in an ice bath for 4 hours to obtain precipitated SNAP crystals. The SNAP crystals appear green in color. After precipitation of the SNAP crystals, the precipitate was collected, and vacuum dried overnight in the dark to remove any trace solvent present with the crystals.

2.3. Fabrication of polymer composites

SNAP-Se polymer composites were prepared by dissolving 70 mg/mL CarboSil in THF at room temperature for 2 h in the dark. After completely dissolving, 10 wt% SNAP was added to the solution and stirred for an additional 10 min. The solution was then cast into a 2.5 cm (diameter) Teflon mold and was allowed to dry overnight in the dark to prevent any premature light-induced NO release. Carbosil control samples were made similarly, but without the addition of SNAP. Each sample was then cut into 1 cm² circular samples and dip coated into a 50 mg/mL solution of CarboSil followed by a second coat of 50 mg/mL CarboSil solution containing different concentrations of Se (1, 5, 10 wt% Se). Control samples were fabricated by coating twice with 50 mg/mL CarboSil without Se. The dip-coated polymer composites were dried overnight.

The following polymer composites were fabricated for the study:

1. 70 mg/mL CarboSil dip coated twice with 50 mg/mL CarboSil (hereon as CarboSil)

2. 10 wt% SNAP-incorporated 70 mg/mL CarboSil dip coated twice with 50 mg/mL CarboSil (hereon as C-SNAP)

3. 10 wt% SNAP-incorporated 70 mg/mL CarboSil dip coated with 50 mg/mL CarboSil and again dip coated with 1 wt% Se-incorporated 50 mg/mL CarboSil (hereon as SNAP-Se-1)

4. 10 wt% SNAP-incorporated 70 mg/mL CarboSil dip coated with 50 mg/mL CarboSil and again dip coated with 5 wt% Se-incorporated 50 mg/mL CarboSil (hereon as SNAP-Se-5)

5. 10 wt% SNAP-incorporated 70 mg/mL CarboSil dip coated with 50 mg/mL CarboSil and again dip coated with 10 wt% Se-incorporated 50 mg/mL CarboSil (hereon as SNAP-Se-10)

6. 70 mg/mL CarboSil dip coated with 50 mg/mL CarboSil and again dip coated with 1 wt% Se-incorporated 50 mg/mL CarboSil (hereon as C-Se)

2.4. Nitric oxide kinetics

NO release:

Nitric oxide release from the samples was measured using Sievers 280i Nitric Oxide Analyzers (NOAs, GE Analytical, Boulder, Colorado, US). The composites were added to reaction vessels containing 3 mL of PBS. The sample holder was partially immersed into a water bath at 37°C to maintain a physiologically relevant temperature. A Pasteur pipet connected to N₂ supply tank was used to seal off the open end of the sample holder as well as carry N₂ gas into the PBS containing the sample. Reaction vessels were kept away from light at all times to avoid unnecessary activation of NO release in the presence of light. Nitric oxide released by the composites (n = 3) were simultaneously swept and purged by high purity N₂ gas at a constant flow rate of 200 mL min⁻¹. The NO purged from the chamber flows into the chemiluminescence reaction chamber. In the reaction chamber, NO reacts with ozone supplied from a separate oxygen tank to produce nitrogen dioxide at an excited state $({\text{NO}}^{2^{\ast }})$. When the nitrogen dioxide decays, it emits a photon which is used to detect the original concentration of NO. Real time measurement of NO was measured in the form of ppb, which was converted to NO flux units by incorporating NOA constants (mol ppb⁻¹ s⁻¹). Prior to measuring the flux, a baseline measurement was first conducted for 1-2 minutes. Samples were incubated at 37°C in the dark in between measurements. The flow rate was set to 200 mL/min with a chamber pressure of 5.4 Torr and an oxygen pressure of 6.0 psi.

NO generation:

NO generated from the Se composites were measured via Sievers 280i NOAs. GSNO, used as the substrate for NO generation was prepared by the reaction of equimolar GSH and NaNO₂ in 0.06 M H₂SO₄. The prepared solution had GSNO (1 μM) and 30 μM GSH in an amber reaction vessel containing PBS at 37 °C. An addition of 0.5 mM EDTA chelated and stabilized the GSNO in the solution to prevent spontaneous generation of NO due to heat.³⁹ After a baseline of release was obtained, a 1 cm² C-Se polymer composite sample was placed in the reaction vessel. For the study involving NO generation measurement after exposure to fibrinogen (Fg), the C-Se samples were pre-adsorbed in Fg from human plasma (prepared at a concentration of 2 mg mL⁻¹ in phosphate buffer) for 60 min. After 60 mins, the samples were washed with phosphate buffer thrice to get rid of non-bound proteins and then placed in the reaction vessel. NO was continuously swept from the headspace of the sample vessel and purged from the solution with a nitrogen sweep gas and bubbler into the chemiluminescence detection chamber.

2.5. Scanning electron microscopy with energy-dispersive x-ray spectroscopy

Scanning Electron Microscopy (SEM, FEI Teneo, FEI Co.) is a technique utilizing a focused electron beam to examine surface morphology. The SEM instrument was accompanied with an Energy-Dispersive X-ray spectroscopy system (EDS, Oxford Instruments) in order to determine the concentration and dispersion of SNAP and Se nanoparticles in the samples using elemental analysis. The presence of SNAP was measured by the detection of sulfur located in the S-NO bond. An accelerating voltage of 5.00 kV and 20.00 kV was employed to examine samples with SEM and EDS, respectively. Samples were coated with 10 nm of gold-palladium using a Leica sputter coater prior to inspection (Leica Microsystems).

2.6. Surface wettability

Surface wettability provides information on the hydrophobicity or hydrophilicity of the surface. The static contact angle was measured for the purpose using Krüss DSA 100 drop shape analyzer. Contact angle was measured for CarboSil, C-SNAP, C-Se and SNAP-Se-1. A 1 μL droplet of water was placed on the samples that were kept on glass slides. The average of left and right contact angles of the water drop on the composite surface were measured via the Krüss software.

2.7. Quantification of SNAP and Se Leaching

2.7.1. Measurement of SNAP leaching

SNAP leaching was measured under physiological conditions for over the course of 24 h. C-SNAP and SNAP-Se-1 polymer composites were soaked 2 mL in PBS at an adjusted pH of 7.4 and incubated before measuring with a Thermo Scientific Genysis 10S UV-Vis Spectrophotometer (UV-Vis). Absorbance was recorded at 340 nm for each sample throughout several timepoints in the ~24 hour timespan, which corresponds to the presence of the S-nitroso group of the SNAP molecule.⁴⁰ PBS was used as a blank control. The concentration of SNAP leaching was calculated from a calibration curve based on known SNAP concentrations dissolved in a PBS solution.

2.7.2. Measurement of Se leaching

Inductively coupled plasma mass spectroscopy (ICP-MS) is a technique used to detect trace elements of interest. During this study, a VG ICP-MS Plasma Quad 3 instrument was used to analyze the samples for Se leaching from the fabricated composites. Samples containing Se interface were soaked in DMEM for 24 h under physiological conditions (37° C, 5% CO₂). The polymer composites were then removed and the leachate solutions were tested for ⁸²Se isotopes by adapting a previously established protocol.⁴¹

2.8. In vitro bacterial adhesion study

To determine the antimicrobial effect of the SNAP-Se, measurement of viable bacteria adhered to the surface of the samples were determined based on a previously established protocol.⁴² Both E. coli (ATCC 11303) and S. aureus (ATCC 6538) were used to assess the antibacterial activity of SNAP-Se-1 against Gram-positive and Gram-negative bacterial strains compared to controls. Isolated strains of E. coli and S. aureus were inoculated in 10 mL of LB broth at 37° C for 14 h at 150 rpm. The culture was then centrifuged for 7 min at 2500 rpm and washed with sterile PBS (pH 7.4). The solution was centrifuged again for 7 min at 2500 rpm and resuspended in PBS by vortexing for 60s. The solution was diluted with PBS to reach a concentration of ~10⁸ CFU/mL. Using a 24 well-plate, each sample (CarboSil, C-Se, C-SNAP and SNAP-Se-1) was incubated with 1 mL of the final bacteria solution for 24 h at 37° C at 150 rpm (n=5).

To evaluate the reduction in viability of bacteria adhered to the polymer composites, they were taken from the 24 well plate after 24 h of exposure and rinsed in PBS to remove any loosely attached bacteria. Adhered bacteria were detached by sonication (Omni-TH sonicator) for 60 s at 25000 rpm and vortexed for an additional 60 s. The resulting solution consisting of the detached bacteria were then serially diluted (10⁻¹ to 10⁻⁵) and plated on LB agar plates for 24 h at 37° C. Colony-forming units (CFU) were counted to determine the number of viable bacteria per cm² of both the control and test samples. The percentage of reduction in bacterial viability was determined by the following equation:

$$\begin{gathered} \%Reductioninbacterialviability=\frac{C_{control}-C_{test}}{C_{control}}x100 \\ \text{where}C=\frac{CFU}{c{m}^2} \end{gathered}$$

Protocols regarding the use of bacteria strains were used in a BSL-2 facility approved by the University of Georgia.

2.9. In vitro platelet adhesion assay

All protocols involving the use of whole blood, plasma, and platelets were approved by the Institutional Animal Care and Use Committee at the University of Georgia prior to experimentation. Fresh porcine blood was drawn through a blind draw with 3.8% sodium citrate (9:1 blood/citrate ratio). After collection, the blood was centrifuged at 300 RCF for 13 minutes and again at 4000 RCF for 20 min using an Allegra X-30 Centrifuge (Beckman-Coulter, Brea, CA) to separate out the platelet rich plasma (PRP) and platelet poor plasma (PPP), respectively. Next, the total platelet count was determined using a hemocytometer (Fisher), and the PRP and PPP were combined to achieve a final concentration of 2 x 10⁸ platelets/mL. CaCl₂ was added at a concentration of 5 mM to reverse the effect of sodium citrate. Samples were added to culture tubes containing 3 mL of the final platelet concentration and incubated at 37°C on a rocker at 25 rpm for 2 h. The samples were then removed and infinitely washed in PBS to remove any loosely attached platelets. Next, an LDH assay was prepared to determine the number of adhered platelets. Samples were stored in Eppendorf tubes containing 500 μL of 2 v/v% Triton-PBS solution for 30 min at room temperature, which lyses the platelets adhered to the surface of the samples. Solutions were plated in a 96-well plate, and Roche Cytotoxicity Detection Kit was utilized to quantify platelet adherence using a BioTek Cytation5 plate reader (BioTek, Winooski, VT) at 492 nm. A calibration curve composed of known platelet counts was generated. The % reduction in platelet adhesion in comparison to control polymer composites were calculated by the following formula:

$$\begin{gathered} \%Reduction=\frac{P_{control}-P_{test}}{P_{control}}x100 \\ \text{where}P=\frac{\#ofplatelets}{c{m}^2} \end{gathered}$$

2.10. In vitro cytotoxicity assay

Interactions between host tissues and any biomaterial requires an evaluation of its biocompatibility during development phase. Due to the reaction of host body in response to a foreign substance in contact, the cellular tolerance level becomes an important factor while maintaining safe levels of biochemical stability and morphology. For this purpose, the biological evaluation of the cells in vitro had been conducted to test if the extracts from our material show any toxicity towards the cells.

Cell culture:

3T3 Mouse fibroblast cells (ATCC 1658) were cultured in 75 cm² T-flask containing DMEM with 4.5 g/L glucose and L-glutamine, 10% FBS, and 1% Pen-Strep, and incubated at 37 °C in a humidified atmosphere with 5% CO₂. The fibroblast cells were trypsinized (0.18% trypsin and 5 mM EDTA) post confluency of ~ 90%. The cells were then seeded into a 96-well plate at a concentration of 5000 cells/mL.

In vitro cytotoxicity assay:

The cytotoxicity assay was conducted in accordance with the ISO 10993 standard using a CCK-8 assay kit. The manufacturer’s (Sigma Aldrich) protocol was followed. The CCK-8 kit utilizes the highly water-soluble tetrazolium salt, WST-8 [2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium monosodium salt], which is reduced by dehydrogenases in live cells to form formazan (an orange-colored product). Formazan can be spectroscopically detected at 450 nm wavelength. Thus, the number of living cells is directly proportional to the amount of formazan dye generated by the dehydrogenases in the cells. All protocols pertaining to the use of mammalian cells were approved by the University of Georgia.

Extract preparation:

Extract was obtained by soaking the polymer composites in DMEM by following the ISO standards (ISO 10993-5:2009 Test for in vitro cytotoxicity). Control and test samples were soaked (after UV sterilization) in DMEM in amber vials (to prevent NO release by light stimulation) and incubated for 24 h at 37 °C. After 24 h, the samples were removed, and the extracts used for the study.

Cell viability:

A 96-well plate containing suspension of 100 μL of cultured cells (5000 cells/mL) in each well was prepared and pre-incubated in a humidified incubator at 37 °C, 5% CO₂ for 24 h. After 24 h, the cells were exposed to 10 μL of the different extracts (n = 7) and incubated for another 24 h to allow the extracts to act on the cells. To each of the wells, 10 μL of the CCK-8 solution was added and incubated for 3 h. The absorbance was measured at 450 nm. Results have been reported as percentage relative cell viability (with respect to positive control) using the following equation.

$$\%Cellviability=\frac{Absorbanceofthetestsamples-Blank}{Absorbanceofthecontrolsamples-Blank}\times 100$$

2.11. Statistical analysis of data

All data are expressed as mean ± standard deviation. The results between the control and test samples were analyzed by a comparison of means using a student’s t-test assuming unequal variances. Values of p < 0.05 were considered statistically significant.

3. Results & Discussion

3.1. Fabrication of combined NOrel/NOgen polymer composites

The objective in design and fabrication of our biomaterial was to integrate NOrel and NOgen chemistries within a single platform in order to promote antibacterial and antiplatelet activity. By doing so, the material will have NO donor reservoir (SNAP) which will provide acute NO release that will prevent the initial onslaught of bacteria and blood clots, as well as have the capability of NOgen materials to generate NO in a highly localized manner from the RSNO present in circulating blood. At the same time, the aim of the study was to minimize complexity in design and fabrication as well as improve functionality. In principle, the design framework with blended Se interface has multiple utility. Apart from being a NOrel material due to the presence of SNAP, the Se interface imparts NOgen capability to the biomaterial which, in practice, can potentially solve the dual problems of infection and thrombogenicity without the limitations of time as a factor.

X-ray diffraction studies in the past have shown that when the concentration of SNAP exceeds its CarboSil solubility (> 4 wt%), the remaining SNAP crystallizes, resulting in longer shelf life and NO release.^43-44 Therefore, based on the previously conducted studies, 10 wt% SNAP was used in all the polymer composites. The interfacial layers had 1, 5, and 10 wt% of Se blended with 50 mg/mL CarboSil. The middle layer consisted only of 50 mg/mL CarboSil for all samples in order to prevent any leaching of SNAP from the base polymer composite. The C-SNAP polymer composite had a thickness of 276.83 ± 3.14 μm while the ones with Se interface, the thickness was 436.013 ± 16.77 μm. The following sections discuss in detail physical and biological characterizations exploring the functionalities stated above.

3.2. Analysis of NO release and generation

3.2.1. In vitro NO-releasing kinetics

The initial evaluation of NO release was the basis of selection of Se weight percentage for the Se interface. Therefore, the polymer composites under consideration were C-SNAP, SNAP-Se-1, SNAP-Se-5 and SNAP-Se-10. It is pertinent to note that the endothelium generates NO at a rate ranging between 0.5 – 4.0 x 10⁻¹⁰ mol cm⁻² min⁻¹, which plays key roles in physiological processes such as vasodilation, angiogenesis, and platelet aggregation and activation.⁴⁵ Hence, to mimic the native endothelium careful modulation of flux from NO-releasing polymer composites is desired. The addition of the Se interface significantly increased the release of NO from the polymer composites on the first day (Figure 1). On Day 1 the C-SNAP had a flux of 2.24 ± 0.3 x 10⁻¹⁰ mol cm⁻² min⁻¹, similar to release rates reported in previous literature.^{42, 46} The SNAP-Se-1, SNAP-Se-5 and SNAP-Se-10 polymer composites showed release rates of 5.89 ± 1.18 x 10⁻¹⁰ mol cm⁻² min⁻¹, 15.70 ± 3.91 x 10⁻¹⁰ mol cm⁻² min⁻¹, and 12.18 ± 3.19 x 10⁻¹⁰ mol cm⁻² min⁻¹, respectively, which were found to be significantly higher than the initial NO release from the C-SNAP polymer composites (P < 0.05). While SNAP-Se-5 and SNAP-Se-10 significantly increased the initial release from the polymer composites compared to C-SNAP and SNAP-Se-1, this trend did not continue during the entire 7-day period. In fact, on Day 7, SNAP-Se-5 and SNAP-Se-10 release rates were 0.23 ± 0.04 x 10⁻¹⁰ mol cm⁻² min⁻¹ and 0.07 ± 0.003 x 10⁻¹⁰ mol cm⁻² min⁻¹, respectively which was similar to the C-SNAP flux of 0.07 ± 0.05 x 10⁻¹⁰ mol cm⁻² min⁻¹. SNAP-Se-1, however, maintained a physiologically relevant flux of 0.8 ± 0.5 x 10⁻¹⁰ mol cm⁻² min⁻¹ by Day 7. Also, interestingly, no significant difference in initial release was found from increasing the concentration of Se beyond 5% (P > 0.05). NO release on Day 1 provides circumstantial evidence suggesting the presence of the Se interface can potentially play a catalytic role. Although the exact mechanism of Se based catalysis is yet to be identified, previous researchers have found other Se-based catalysts to have similar effects on the NO release.^{32-33, 35} High concentrations of NO can alter cellular functions, which can result in cytotoxicity towards mammalian cells.⁴⁷ Moreover, modifiable catalytic surface technologies which successfully increase the NO release for an extended period of time can increase the antimicrobial and thromboresistant activity of NO devices while simultaneously avoiding cytotoxicity towards mammalian cells. Therefore, due to the initial higher release and higher concentrations of Se in SNAP-Se-5 and SNAP-Se-10, we proceeded with SNAP-Se-1 to maximize antimicrobial and antiplatelet activity while minimizing the risk of cytotoxicity. For the duration of the study, SNAP-Se-1 polymer composites have been subjected to further evaluation due to its ability to maximize NO flux and longevity while minimizing Se required.

Figure 1:

Graph reporting the NO flux analysis of SNAP, SNAP-Se-1, SNAP-Se-5 and SNAP-Se-10 polymer composites. Error bars represent standard deviation (n=3). Standard two-tailed student t-test was performed to determine the significance groups at the same time points. ‘*’= p < 0.05 vs. control; ‘#’ = p < 0.05 vs. SNAP-Se-1; ‘$’ = p < 0.05 vs. SNAP-Se-5; ‘%’ = p < 0.05 vs. SNAP-Se-10.

3.2.2. In vitro NO generation

The primary limitation for NOrel biomaterials that incorporate NO donating molecules is that this reservoir eventually becomes exhausted over a period of time. This has led to the investigation of NO generating materials that are able to catalytically react with endogenous RSNOs, such as GSNO, that are present within the blood to locally release NO at the material interface. The catalytic release of RSNOs has been demonstrated in the past using this mechanism from polymers containing certain metals such as copper or metal organic framework (MOF) molecules.^48-49 Polymer films containing organoselenium moieties have also been proven to catalyze RSNO decomposition to generate NO.³³ However, catalytic generation of NO from polymer composites with embedded Se placed in solutions with low concentrations (1 μM) of RSNOs have not been investigated. This concentration corresponds to the approximate range of GSNO present in blood (0.5 – 2 μM)⁵⁰ that has the potential to be catalytically influenced from the Se containing materials. Passive NO generation was tested by submerging the fabricated C-Se polymer composites into a PBS solution containing 1 μM GSNO and 30 μM GSH to simulate the RSNO environment seen in the blood. Additionally, 500 μM EDTA was incorporated into the solution to chelate any possible trace metals and to ensure the NO generation was solely from the immersed C-Se. The C-Se interfaces are capable of generating NO from endogenous RSNOs (Figure 2). The polymer composites were able to generate a flux (> 0.5 x 10⁻¹⁰ mol cm⁻² min⁻¹) capable of mimicking the protective effects seen in native endothelium. To simulate better physiological condition, the C-Se polymer composites were also pre-adsorbed with Fg (2 mg mL⁻¹ in phosphate buffer for 1 h. As expected, the NO generation levels were lower (~0.3 x 10⁻¹⁰ mol cm⁻² min⁻¹) but the polymer composites were still able to generate a physiologically relevant amount of NO. While the polymer composites that contain both SNAP and Se exhibit high initial releases of NO for the few first days, which is necessary for eliminating bacteria and preventing biofilm formation, this subsequent prolonged passive NO generation still exhibits levels of NO that is capable of preventing platelet activation. This unique combination of NOrel and NOgen chemistries within medical devices polymers can be immensely advantageous in terms of overcoming challenges with device implants by providing potent antimicrobial activity both acutely (at the time of surgical placement) as well as continued NO generation to protect from chronic or late device infections.

Figure 2:

Graph demonstrating representative real-time NO generation from C-Se (containing 1% Se) in the presence of GSNO (1 μM) before and after exposure to fibrinogen under physiological conditions (37 °C, pH 7.4)

3.3. Analysis of surface composition and morphology with SEM-EDS

SEM was deployed to investigate the surface morphology of the SNAP-Se-1 polymer composite. Prior to adding a Se interfacial layer, the SNAP incorporated CarboSil was imaged (Figure 3A). The addition of a Se interface (Figure 3C) did not significantly alter the surface morphology of the original composite. To confirm the presence of SNAP incorporated into the CarboSil composites and Se incorporated Carbo1Sil on the surface, elemental maps were constructed using EDS. In order to detect SNAP, the presence of sulfur (S) located in the S-NO bond of SNAP was determined (Figure 3B). After coating, the SNAP-Se-1 polymer composite was scanned to determine the presence of Se in the outer layer (Figure 3D). Both S and Se maps indicated the presence and even distribution of SNAP and Se in their respective layers.

Figure 3 –

SEM and EDS analysis of SNAP-Se-1 polymer composites. (A) – SEM image of SNAP composite prior to being coated with Se layer and (C) - after coating with 1% Se interface. (B) - Prior to coating, the detection of sulfur was used to analyze the presence of SNAP in the SNAP composite. (D) -After coating, the detection of Se was used to analyze the presence of Se in the SNAP-Se-1 polymer composite.

3.4. Leaching of SNAP and Se

3.4.1. Leaching of SNAP

The initial leaching of NO donors can affect the lifetime of NO-releasing medical devices by significantly reducing the time duration of NO release from devices. Additionally, devices can exhibit a “burst release” effect, resulting in an elevated NO flux during the first few hours after beginning its use. Controlling excessive leaching is crucial as the amount of NO released into the bloodstream can cause adverse side effects such as vasodilation, resulting in lowered blood pressure.⁴⁰ In order to avert this, hydrophobic coatings have previously been applied to control the amount of NO donor leached, especially in the first few hours of use.⁵¹ Additional layers of CarboSil have in the past been optimized to result in less SNAP leaching when samples are initially immersed.⁴² Therefore, to reduce SNAP leaching, each sample was coated with two hydrophobic polymer layers (CarboSil and/or Carbosil-Se). Because Se species have shown to have antimicrobial effects by directly interacting with bacteria, the outer most layer from SNAP-Se-1 and C-Se polymer composites were composed of 1 wt% (w/w) of Se incorporated CarboSil acting as the interface, while the middle coat consisted of only CarboSil.^52-54 To determine the amount of SNAP leaching from the polymer composites, samples were soaked in PBS and measured at different time points over a 24 h time period with UV-Vis spectrophotometer. Both the C-SNAP and SNAP-Se-1 polymer composites exhibited similar leaching patterns, only leaching a total of 0.1990 mg/cm² and 0.1469 mg/cm² of SNAP per cm² of polymer composite after over 24 hours of incubation in PBS, respectively (Figure 4A). This suggests that the addition of Se to the polymer composites did not significantly affect SNAP leaching. Less than 5% of the total NO donor stored in each of the samples were leached after 24 h of incubation in an aqueous environment, suggesting that the amount of SNAP leached will not significantly affect the lifetime of NO release from the composites (Figure 4B).

Figure 4 –

SNAP leaching (A) and % of SNAP remaining (B) in the polymer composites after 24 h. Both C-SNAP and SNAP-Se-1 showed less than 5% of SNAP leaching from the composites. No significant difference was found between either sample. The data is reported in means ± SD.

3.4.2. Leaching of Se

Although Se is essential for survival, excess Se could lead to cytotoxic effects. Previous research has suggested that an intake level of 400 μg/day of Se is the upper tolerable limit for an individual.⁵⁵ Therefore, when considering the possible side effects of cytotoxicity, it is important to ensure that the level of Se leaching is below the tolerable threshold. However, low levels of Se leaching can be promising. Se in the past have demonstrated superior bactericidal capabilities.³⁷ In order to determine the level of Se leaching from the composites, a VG ICP-MS Plasma Quad 3 instrument was used to measure 24 h leachate samples in complete DMEM cell culture media. Both C-Se and SNAP-Se-1 polymer composites leached less than 6 μg/L Se into the media, which equates to roughly 0.01% of total Se originally incorporated into the polymer composites. This is significantly less than the suggested daily Se concentration intake. There was no significant difference found between the leaching of SNAP-Se-1 and C-Se (Table 1).

Table 1 –

Se leaching measurements of C-Se and SNAP-Se-1 polymer composites. No statistical difference of Se leaching was found between the SNAP-Se-1 and C-Se polymer composites. Data is reported in means ± SD (n=3).

Polymer composite	Se leaching (μg/L)	% Se leaching
C-Se	5.2 ± 1.2	0.014 ± .003
SNAP-Se-1	4.4 ± 1.3	0.012 ± .003

3.5. Surface wettability

Surface wettability provides information about the surface-liquid interfacial tension by establishing the angle of a liquid drop on the solid surface. The surface property of polymer composites can dictate the interaction of bacteria and blood proteins with the polymer composites.^56-57 A surface contact angle higher than 90° is generally considered to be hydrophobic.⁵⁸ As shown in Table 2, the polymer composites containing the Se interface had a slightly lower hydrophobicity. The result is also in agreement with reduction in water contact angle observed on a Se coated titanium substrate which was mainly ascribed to contribution of the roughness factor and presence of air pockets.⁵⁵

Table 2 -.

Water contact angle measurements of the polymer composites using Krüss DSA100 Drop Shape Analyzer. Data represents mean ± SD (n=3)

Polymer composite	Contact Angle (°)
CarboSil	100.12 ± 0.57
C-Se	88.98 ± 1.30
C-SNAP	100.00 ± 1.44
SNAP-Se-1	80.28 ± 0.94

3.6. Reduction in adhered bacteria

Infection is problematic for the use of medical devices, limiting success and rates of mortality. Unlike antibiotics, which have specific mechanisms to kill bacteria, NO has a broad range of bactericidal mechanisms and therefore has encountered less resistance than antibiotics.²⁰ The first few hours after device implementation are crucial in preventing infection, as medical device-related infection due to biofilm formation has been previously reported to occur rapidly after insertion (less than 24 h).^{2, 59} In order to avert the risk of infection, NO-incorporated devices have been devised to reduce the number of viable bacteria adhered to the surface of devices.^{21, 60} In addition, Se has been shown to have antimicrobial behavior.^61-62 In this study, we set out to explore the antibacterial effect of incorporating Se with the NO donor SNAP in the hydrophobic polymer CarboSil. In order to do so, the polymer composites were exposed to two strains of bacteria commonly associated with HAIs, S. aureus (Gram-positive) and E. coli (Gram-negative) for 24 h to assess the reduction in the viability of bacteria adhered to the surface of the polymer composites. The combination of Se and SNAP best reduces the number of bacteria adhered to the surface of the polymer composites for both strains of bacteria (Figure 5).

Figure 5-.

Graph demonstrating the reduction in the viability of adhered Gram-negative E. coli and Gram-positive S. aureus after 24 h. Error bars represent standard deviation (n=5). ‘*’= p < 0.05 vs. CarboSil controls.

The combination of Se interface with NO release was able to reduce the number of adhered E. coli and S. aureus by 99.18 ± 0.13% and 99.65 ± 0.28% respectively (Table 3). The reduction efficiency of SNAP-Se-1 towards viability of adhered bacteria can be attributed to the compounded antimicrobial activity of both agents as well as the increase in NO flux in 24 h. The bactericidal effect of NO includes multiple mechanism routes including DNA cleavage, nitrosative and oxidative action, and formation of peroxynitrite or superoxide.^{20, 22} In addition, Se had been shown to inhibit growth of bacteria by impairing DNA structures or by reactive oxygen species (ROS) generation in several studies.^{54, 63} The exact mechanism, however, is yet to be established. Due to the ability of Se interface to generate physiologically relevant NO catalytically in the presence of RSNOs which circulate in blood, it would, in principle, prove to be effective in resisting chronic infections locally. The current study can form the basis of further assessments into mechanism and surface-associated catalytic NO generation and its ability to prevent bloodstream infections in vivo.

Table 3-.

Comparative analysis of bacterial CFU/cm² adhering to CarboSil, C-Se, C-SNAP and SNAP-Se-1 polymer composites.

Bacterial strain		CarboSil	C-Se	C-SNAP	SNAP-Se-1
S. aureus	Average CFU	9 × 106	2.43× 106	7.5× 104	3.17× 104
	Reduction efficiency (%)	-	73.01 ± 21.3	99.16 ± 0.27	99.65 ± 0.28
	p-value vs. CarboSil control	-	0.008	0.003	0.003
E. coli	Average CFU	6.6 × 107	2.5× 107	1.33× 107	5.5× 105
	Reduction efficiency (%)	-	62.50 ± 15	80.00 ± 11.4	99.18 ± 0.13
	p-value vs. CarboSil control	-	0.03	0.01	0.005

3.7. Reduction in platelet adhesion

The clinical standard to prevent blood clot formation on indwelling blood-contacting medical devices is systemic anti-platelet and anticoagulation therapies. Unfortunately, systemic anti-platelet and anticoagulation therapies can lead to a number of undesired side effects such as undesired or uncontrolled bleeding. Therefore, a device that can exhibit inherent antiplatelet activity to reduce the occurrence of thrombosis is of significant interest. In this study, the ability of the polymer composites to prevent platelet adhesion was assessed using an LDH assay after exposure to a porcine platelet concentration of 2 x 10⁸ platelets/mL. The combination of Se interface and SNAP best inhibited platelets from adhering, showing an 85.5% reduction as compared to CarboSil controls (Figure 6).

Figure 6-.

Graphical representation of the LDH assay demonstrating reduction in platelet adhesion after exposure to porcine PRP in a 2 h study. The SNAP-Se-1 polymer composite provided the largest decrease in platelet adhesion, showing 85.5 % reduction when compared to that of the controls. Error bars represent standard deviation. ‘*’= p < 0.05 vs. CarboSil control; ‘#’= p < 0.05 vs. C-SNAP.

From the NO release profiles, it is evident that the SNAP-Se-1 polymer composites exhibited greater NO release compared to C-SNAP. The elevated NO at the upper physiological ranges can significantly inhibit platelet adhesion as observed in previous studies in vitro and in vivo.^{42, 64} It is also interesting to observe a 72.8% reduction in platelet adhesion by C-Se as compared to CarboSil polymer composites. Although the mechanism of inhibition of platelet adhesion or resistance to platelet aggregation by Se is poorly understood, various forms of Se have been shown to reduce or inhibit platelet aggregation.^65-66 Similar reduction in platelet adhesion was observed via SEM study conducted on Se blended polyurethanes.³⁶ The low levels of NO generated from RSNOs as seen from NO generation studies also holds the potential for SNAP-Se-1 polymer composite to be ideal for blood contacting applications (e.g., vascular grafts, stents). In fact, Major et al. demonstrated that in a 4 h rabbit model, extracorporeal loops coated with only 10 wt% Cu infused with saline solution showed improved platelet counts compared to control loops with intravenous infusion of an RSNO.²⁷ Moreover, in a study on electrochemically modulated NO generation via a Cu(II)-tri(2-pyridylmethyl)amine (Cu(II)TPMA) catalyst, researchers showed that NOGen materials were capable of reducing thrombus formation in catheters.⁶⁷ In the current scope of this study, further tunability of Se interfacial layer and its interaction with RSNOs embedded in the polymer as well as circulating in blood has not been focused due to limited knowledge of the mechanism through which it reduces platelet adhesion. However, the present result demonstrates that a Se interface on a NO releasing polymer composite can significantly reduce platelet adhesion. This capability can further be extended in combination with other antiplatelet therapies to improve overall hemocompatibility of blood-contacting devices.

3.8. In vitro cytotoxicity

In order to demonstrate the effect of toxicity elicited by extracts from CarboSil, C-SNAP, C-Se and SNAP-Se-1 polymer composites, cytotoxicity testing was conducted in accordance with ISO 10993 standards. NIH 3T3 mouse fibroblast cell line (ATCC 1658) was used for this study. All the polymer composites were soaked in DMEM at 37°C for 24 h to allow sample extracts to diffuse into the medium. After 24 h, the extracts were exposed to mouse fibroblast cells cultured simultaneously with extract preparation. After 24 h of incubation post addition of the extracts, CCK-8 dye was added to each well and incubated for 3 h. The change in color intensity due to formation of formazan was measured at 450 nm using a multiplate reader. The control had no extracts. It consisted of only cells growing on the 96-wells. With respect to the control, the mouse fibroblast cells demonstrated no significant difference in the presence of the extracts from all the samples (n=7, Figure 7). Previous studies conducted by the Handa lab and other labs have shown the non-cytotoxicity and potential biocompatibility of NO releasing polymers via SNAP donor.^{42, 51, 68} In the recent past, cytotoxicity studies conducted with increasing amount of Se in a Se blended polyurethane has shown slight decrease in viability of mouse fibroblast cells. Toxicity elicited by Se or Se species has mostly been ascribed to oxidative damage.⁶⁹ At 1 wt% Se concentration and negligible leaching of Se from the interfacial layer, the viability of mouse fibroblast cells were 91.3 ± 5.4 %. SNAP-Se-1, however, had improved viability at 97.37 ± 4.94 %. The results from this study provides supporting evidence towards the potential biocompatibility of the SNAP-Se-1 polymer composite. Even though, antibacterial and anti-platelet adhesion characteristic of SNAP-Se-1 is of utmost importance, it should not compromise the biocompatibility of mammalian cells. Further in vivo testing in animal models would help establish the biocompatibility of SNAP-Se-1.

Figure 7-.

Graph demonstrating 24 h cell viability of NIH 3T3 mouse fibroblast cells using WST dye based CCK-8 assay. The error bar represents standard deviation (n=7). The results were not statistically significant (p<0.05).

4. Conclusion

The current premise of these findings presents a promising biomaterial specifically designed to integrate properties of both NOrel and NOgen materials in order to achieve localized NO at polymer interfaces. The findings demonstrate that a Se interface containing 1 wt % Se blended with CarboSil on a NO releasing polymer composite elevates the NO release to 5.89 ± 1.18 x 10⁻¹⁰ mol cm⁻² min⁻¹ compared to 2.24 ± 0.36 x 10⁻¹⁰ mol cm⁻² min⁻¹. In addition to that, NO (> 0.5 x 10⁻¹⁰ mol cm⁻² min⁻¹) was also generated by C-Se polymer composites in the presence of 1 μM GSNO thus exhibiting NO generating capabilities. As an extension, in principle, not only does SNAP-Se-1 polymer composites have the much-needed elevated release in the beginning required to prevent infection and platelet adhesion, these surfaces can sustainably generate NO by reacting with RSNOs circulating within the blood. The SNAP-Se-1 significantly reduced the adhesion of E. coli and S. aureus by greater than 2 log reduction which are among prominent pathogens causing nosocomial infections. The LDH assay-based platelet adhesion studies reveal that SNAP-Se-1 can effectively reduce platelet adhesion. Leaching of SNAP and Se was measured through a spectrophotometric method and ICP-MS, respectively. Leaching from either species did not elicit cytotoxicity in mouse fibroblast cells. Both the leaching and cytotoxicity studies provide important supporting evidence towards the biocompatibility of SNAP-Se-1. The results from this work provides proof of principle for further design and development of polymeric biomaterials incorporating both NOrel and NOgen capabilities in a single platform. This unique approach of combining NOrel and NOgen polymer chemistries can provide localized NO at the polymer interface in order to overcome both acute and chronic biocompatibility and microbial challenges associated with indwelling medical devices. However, further testing in pre-clinical settings will be critical to establish antibacterial and anti-platelet adhesion efficacies as well as determine the safety towards clinical use.